摘要
针对振动信号采集过程中混入噪声无法准确判断信息的问题,提出基于集合经验模态的小波半软阈值降噪法。该方法利用EEMD算法的自适应性且能避免IMF(Intrinsic Mode Function)分量混叠的特点以及小波半软阈值函数连续性好且避免重构信号存在偏差的优点,在一定程度上避免了EMD分解过程中出现混叠的现象,且降低了重构信号的偏差,对信号降噪的同时能更准确地将信号波形复原。分别利用仿真信号和实验信号验证该方法的有效性,分析结果表明,基于EEMD的小波半软阈值降噪方法具备两种方法的优点,能够很好地抑制信号中的噪声,并且在重构过程中能较好地复原信号的有用信息。
In view of the vibration signal acquisition process inevitably mixed with noise,affecting the accurate judgment of the information,a wavelet semi-soft threshold denoising method based on ensemble empirical mode decomposition was proposed.The method used the self-adaptability of the EEMD algorithm avoiding the IMF(Intrinsic Mode Function)component aliasing,and it had the advantages of good continuity of the wavelet semi-soft threshold function,which avoided the deviation of the reconstructed signal.To a certain extent,the phenomenon of aliasing in the EMD decomposition process was avoided,and the deviation of the reconstructed signal was reduced,and the signal could be recovered more accurately.The validity of this method was respectively verified by using simulated signal and experimental signal.The analysis results showed that the wavelet semi-soft threshold denoising method based on EEMD could suppress the noise in the signal well,recover useful signal information better in the reconstruction process.
作者
甄龙信
王云龙
邓小艳
张伟锟
ZHEN Longxin;WANG Yunlong;DENG Xiaoyan;ZHANG Weikun(School of Vehicle and Energy,Yanshan University,Qinhuangdao 066000,China)
出处
《探测与控制学报》
CSCD
北大核心
2018年第5期53-57,共5页
Journal of Detection & Control
关键词
集合经验模态分解
小波
阈值函数
降噪
ensemble empirical mode decomposition
wavelet
threshold function
denoising
